Performance during Overclocking
Besides the tests in nominal mode, we would also like to compare the mainboards’ performance in overclocked systems. The thing is that relative performance of overclocker platforms is very often different from what we see in nominal mode.
For the second round of tests we decided to set the FSB frequency at 450MHz. We used the same Core 2 Duo E6850 processor overclocked to 3.6GHz set as 8 x 450MHz. The processor Vcore was increased to 1.45V to ensure better stability. DDR2 memory was also sped up to 1125MHz and the timings were set at 5-5-5-15.
First let me share some good results. Our tests revealed that Gigabyte engineers managed to resolve the issues that we have previously observed during our overclocking experiments. Gigabyte P35-DQ6 mainboard with the BIOS F6 version (just like other GA-P35 mainboards with September BIOS versions) proved stable at higher memory speeds and doesn’t fall behind other mainboards during overclocking any more.
By the way, in this respect I would like to stress that Asus and Gigabyte mainboards are a priori more suitable for overclocker platforms than Foxconn Mars. While the Foxconn mainboard is just beginning to settle in the market, the solutions from other mainboard makers have been selling for a few months already. It means that they have already got rid of most their “early stage issues”, while Foxconn engineers will be just starting on the error correction path.
For example, they will need to do something about the not quite efficient cooling of the chipset North Bridge, which has a negative effect on CPU and memory overclocking on Foxconn Mars. In order to get our Foxconn Mars based system to run at 450MHz FSB and 1125MHz memory frequency, we had to spend a while before we finally found the North Bridge voltage when the mainboard remained stable and the North Bridge didn’t overheat. BY the way, this setting was 1.375V: even with an additional fan on top of Cool Pipe system you can’t push it to 1.5-1.7V like on other mainboards.
Memory Subsystem Synthetic Benchmarks